← LLM Models#1
Claude Opus 4.7
Anthropic
Most capable Claude. Top SWE-Bench score among all models. Complex analysis.
Official page →📖Context200K tokens
🚀Speed50 tok/s
💵Input price$5.00/1M
💸Output price$25.00/1M
Benchmarks
Gray bar = dataset averageMMLU57 academic subjects
90.1%avg 81.8
GPQA DiamondPhD-level science questions
74.3%avg 50.4
HumanEvalPython code generation
92.1%avg 79.9
SWE-Bench VerifiedReal GitHub engineering tasks
72.5%avg 32.4
Composite ScoreScore = MMLU×20% + GPQA×30% + HumanEval×25% + SWE-Bench×25%81.5
More from Anthropic
API Providers
Details
- Provider
- Anthropic
- Parameters
- -
- Context
- 200K tokens
- Speed
- 50 tok/s
- Open Source
- No
- License
- Proprietary
- Released
- Apr 2026